|
QLoRA resource reality check with SGLang
|
09-23-2026, 03:47 PM
What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at HBM constraints. Once you know the plain path works, then add the faster path back in piece by piece.
throughput matters, but tail latency matters too
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
QLoRA resource reality check with SGLang - by deepseeked - 09-22-2026, 01:34 AM
Re: QLoRA resource reality check with SGLang - by batchsize - 09-23-2026, 02:09 AM
Re: QLoRA resource reality check with SGLang - by batchsize - 09-23-2026, 03:47 PM
Re: QLoRA resource reality check with SGLang - by batchpeak - 09-23-2026, 02:56 PM
Re: QLoRA resource reality check with SGLang - by routearch - 09-24-2026, 08:22 AM
Re: QLoRA resource reality check with SGLang - by benchplease - 09-24-2026, 03:55 AM
Re: QLoRA resource reality check with SGLang - by tensorfield88 - 09-24-2026, 12:44 PM
|
